Upload fine-tuned model
Browse files- config.json +108 -7
- generation_config.json +1 -1
- model-00001-of-00004.safetensors +1 -1
- model-00002-of-00004.safetensors +1 -1
- model-00003-of-00004.safetensors +1 -1
- model-00004-of-00004.safetensors +1 -1
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -7
- tokenizer_config.json +1 -1
- trainer_state.json +1405 -0
- training_args.bin +3 -0
config.json
CHANGED
|
@@ -1,13 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "microsoft/Phi-3-mini-4k-instruct",
|
| 3 |
"architectures": [
|
| 4 |
"Phi3ForCausalLM"
|
| 5 |
],
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"auto_map": {
|
| 9 |
-
"AutoConfig": "microsoft/Phi-3-mini-
|
| 10 |
-
"AutoModelForCausalLM": "microsoft/Phi-3-mini-
|
| 11 |
},
|
| 12 |
"bos_token_id": 1,
|
| 13 |
"embd_pdrop": 0.0,
|
|
@@ -16,7 +15,7 @@
|
|
| 16 |
"hidden_size": 3072,
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
"intermediate_size": 8192,
|
| 19 |
-
"max_position_embeddings":
|
| 20 |
"model_type": "phi3",
|
| 21 |
"num_attention_heads": 32,
|
| 22 |
"num_hidden_layers": 32,
|
|
@@ -25,12 +24,114 @@
|
|
| 25 |
"pad_token_id": 32000,
|
| 26 |
"resid_pdrop": 0.0,
|
| 27 |
"rms_norm_eps": 1e-05,
|
| 28 |
-
"rope_scaling":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"rope_theta": 10000.0,
|
| 30 |
-
"sliding_window":
|
| 31 |
"tie_word_embeddings": false,
|
| 32 |
"torch_dtype": "float32",
|
| 33 |
-
"transformers_version": "4.
|
| 34 |
"use_cache": true,
|
| 35 |
"vocab_size": 32064
|
| 36 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"Phi3ForCausalLM"
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
"auto_map": {
|
| 8 |
+
"AutoConfig": "microsoft/Phi-3-mini-128k-instruct--configuration_phi3.Phi3Config",
|
| 9 |
+
"AutoModelForCausalLM": "microsoft/Phi-3-mini-128k-instruct--modeling_phi3.Phi3ForCausalLM"
|
| 10 |
},
|
| 11 |
"bos_token_id": 1,
|
| 12 |
"embd_pdrop": 0.0,
|
|
|
|
| 15 |
"hidden_size": 3072,
|
| 16 |
"initializer_range": 0.02,
|
| 17 |
"intermediate_size": 8192,
|
| 18 |
+
"max_position_embeddings": 131072,
|
| 19 |
"model_type": "phi3",
|
| 20 |
"num_attention_heads": 32,
|
| 21 |
"num_hidden_layers": 32,
|
|
|
|
| 24 |
"pad_token_id": 32000,
|
| 25 |
"resid_pdrop": 0.0,
|
| 26 |
"rms_norm_eps": 1e-05,
|
| 27 |
+
"rope_scaling": {
|
| 28 |
+
"long_factor": [
|
| 29 |
+
1.0700000524520874,
|
| 30 |
+
1.1200000047683716,
|
| 31 |
+
1.149999976158142,
|
| 32 |
+
1.4199999570846558,
|
| 33 |
+
1.5699999332427979,
|
| 34 |
+
1.7999999523162842,
|
| 35 |
+
2.129999876022339,
|
| 36 |
+
2.129999876022339,
|
| 37 |
+
3.009999990463257,
|
| 38 |
+
5.910000324249268,
|
| 39 |
+
6.950000286102295,
|
| 40 |
+
9.070000648498535,
|
| 41 |
+
9.930000305175781,
|
| 42 |
+
10.710000038146973,
|
| 43 |
+
11.130000114440918,
|
| 44 |
+
14.609999656677246,
|
| 45 |
+
15.409998893737793,
|
| 46 |
+
19.809999465942383,
|
| 47 |
+
37.279998779296875,
|
| 48 |
+
38.279998779296875,
|
| 49 |
+
38.599998474121094,
|
| 50 |
+
40.12000274658203,
|
| 51 |
+
46.20000457763672,
|
| 52 |
+
50.940006256103516,
|
| 53 |
+
53.66000747680664,
|
| 54 |
+
54.9373893737793,
|
| 55 |
+
56.89738845825195,
|
| 56 |
+
57.28738784790039,
|
| 57 |
+
59.98738479614258,
|
| 58 |
+
60.86738586425781,
|
| 59 |
+
60.887386322021484,
|
| 60 |
+
61.71739196777344,
|
| 61 |
+
62.91739273071289,
|
| 62 |
+
62.957393646240234,
|
| 63 |
+
63.41739273071289,
|
| 64 |
+
63.8173942565918,
|
| 65 |
+
63.83739471435547,
|
| 66 |
+
63.897396087646484,
|
| 67 |
+
63.93739700317383,
|
| 68 |
+
64.06739807128906,
|
| 69 |
+
64.11434936523438,
|
| 70 |
+
64.12435150146484,
|
| 71 |
+
64.15435028076172,
|
| 72 |
+
64.19435119628906,
|
| 73 |
+
64.24435424804688,
|
| 74 |
+
64.57435607910156,
|
| 75 |
+
64.69000244140625,
|
| 76 |
+
64.76000213623047
|
| 77 |
+
],
|
| 78 |
+
"short_factor": [
|
| 79 |
+
1.1,
|
| 80 |
+
1.1,
|
| 81 |
+
1.1,
|
| 82 |
+
1.3000000000000003,
|
| 83 |
+
1.3500000000000003,
|
| 84 |
+
1.3500000000000003,
|
| 85 |
+
1.4000000000000004,
|
| 86 |
+
1.5500000000000005,
|
| 87 |
+
2.000000000000001,
|
| 88 |
+
2.000000000000001,
|
| 89 |
+
2.000000000000001,
|
| 90 |
+
2.000000000000001,
|
| 91 |
+
2.000000000000001,
|
| 92 |
+
2.000000000000001,
|
| 93 |
+
2.000000000000001,
|
| 94 |
+
2.000000000000001,
|
| 95 |
+
2.000000000000001,
|
| 96 |
+
2.000000000000001,
|
| 97 |
+
2.000000000000001,
|
| 98 |
+
2.000000000000001,
|
| 99 |
+
2.000000000000001,
|
| 100 |
+
2.000000000000001,
|
| 101 |
+
2.000000000000001,
|
| 102 |
+
2.000000000000001,
|
| 103 |
+
2.000000000000001,
|
| 104 |
+
2.0500000000000007,
|
| 105 |
+
2.0500000000000007,
|
| 106 |
+
2.0500000000000007,
|
| 107 |
+
2.0500000000000007,
|
| 108 |
+
2.0500000000000007,
|
| 109 |
+
2.0500000000000007,
|
| 110 |
+
2.1000000000000005,
|
| 111 |
+
2.1000000000000005,
|
| 112 |
+
2.1500000000000004,
|
| 113 |
+
2.25,
|
| 114 |
+
2.25,
|
| 115 |
+
2.25,
|
| 116 |
+
2.25,
|
| 117 |
+
2.25,
|
| 118 |
+
2.3999999999999995,
|
| 119 |
+
2.4499999999999993,
|
| 120 |
+
2.499999999999999,
|
| 121 |
+
2.6999999999999984,
|
| 122 |
+
2.6999999999999984,
|
| 123 |
+
2.7499999999999982,
|
| 124 |
+
2.799999999999998,
|
| 125 |
+
2.8999999999999977,
|
| 126 |
+
3.049999999999997
|
| 127 |
+
],
|
| 128 |
+
"type": "longrope"
|
| 129 |
+
},
|
| 130 |
"rope_theta": 10000.0,
|
| 131 |
+
"sliding_window": 262144,
|
| 132 |
"tie_word_embeddings": false,
|
| 133 |
"torch_dtype": "float32",
|
| 134 |
+
"transformers_version": "4.51.3",
|
| 135 |
"use_cache": true,
|
| 136 |
"vocab_size": 32064
|
| 137 |
}
|
generation_config.json
CHANGED
|
@@ -7,5 +7,5 @@
|
|
| 7 |
32007
|
| 8 |
],
|
| 9 |
"pad_token_id": 32000,
|
| 10 |
-
"transformers_version": "4.
|
| 11 |
}
|
|
|
|
| 7 |
32007
|
| 8 |
],
|
| 9 |
"pad_token_id": 32000,
|
| 10 |
+
"transformers_version": "4.51.3"
|
| 11 |
}
|
model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4961852416
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80090c0ec5dfe7066377a3cc545ac216fd6741d8d88895f37ecff26a25a72237
|
| 3 |
size 4961852416
|
model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4983111176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0656a1cc2e04850b6ef2963e0250cd620faecdbf88b58071d4c3757934aa8a43
|
| 3 |
size 4983111176
|
model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4945374704
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16511ee7337aa20595cd6e4025396bb209c1e34dee938838ec739fb48a7dcf0a
|
| 3 |
size 4945374704
|
model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 394002560
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b900f38794888826263302802614491ebb47c32f22ddc42bbb09e072e8680f3c
|
| 3 |
size 394002560
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3bd72cb4c0c60b6efcc4dc7d4e981861f09b6c5eae7f57d652c74815909a54bf
|
| 3 |
+
size 7762960962
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:285abb08886d14c329c3145cd21092c92a575d4647731c4c6bacdc52838df314
|
| 3 |
+
size 14244
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34d01644342e74e4c6b26bcf6157d30b6c7b4ae37f6998df0a4f5e710352b706
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42ac1193d9afd5e9e6a069954ba46bddc93f719e87099011fc640726439b771c
|
| 3 |
+
size 1064
|
special_tokens_map.json
CHANGED
|
@@ -13,13 +13,7 @@
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
-
"pad_token":
|
| 17 |
-
"content": "<|endoftext|>",
|
| 18 |
-
"lstrip": false,
|
| 19 |
-
"normalized": false,
|
| 20 |
-
"rstrip": false,
|
| 21 |
-
"single_word": false
|
| 22 |
-
},
|
| 23 |
"unk_token": {
|
| 24 |
"content": "<unk>",
|
| 25 |
"lstrip": false,
|
|
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
+
"pad_token": "<|endoftext|>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
"unk_token": {
|
| 18 |
"content": "<unk>",
|
| 19 |
"lstrip": false,
|
tokenizer_config.json
CHANGED
|
@@ -122,7 +122,7 @@
|
|
| 122 |
"eos_token": "<|endoftext|>",
|
| 123 |
"extra_special_tokens": {},
|
| 124 |
"legacy": false,
|
| 125 |
-
"model_max_length":
|
| 126 |
"pad_token": "<|endoftext|>",
|
| 127 |
"padding_side": "left",
|
| 128 |
"sp_model_kwargs": {},
|
|
|
|
| 122 |
"eos_token": "<|endoftext|>",
|
| 123 |
"extra_special_tokens": {},
|
| 124 |
"legacy": false,
|
| 125 |
+
"model_max_length": 131072,
|
| 126 |
"pad_token": "<|endoftext|>",
|
| 127 |
"padding_side": "left",
|
| 128 |
"sp_model_kwargs": {},
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1405 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 10,
|
| 7 |
+
"global_step": 625,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008,
|
| 14 |
+
"grad_norm": 15.360989570617676,
|
| 15 |
+
"learning_rate": 8.000000000000001e-06,
|
| 16 |
+
"loss": 5.9088,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.016,
|
| 21 |
+
"grad_norm": 7.644533634185791,
|
| 22 |
+
"learning_rate": 1.8e-05,
|
| 23 |
+
"loss": 4.5305,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.016,
|
| 28 |
+
"eval_loss": 4.309304714202881,
|
| 29 |
+
"eval_runtime": 110.5075,
|
| 30 |
+
"eval_samples_per_second": 4.525,
|
| 31 |
+
"eval_steps_per_second": 4.525,
|
| 32 |
+
"step": 10
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.024,
|
| 36 |
+
"grad_norm": 13.015691757202148,
|
| 37 |
+
"learning_rate": 1.999791251006346e-05,
|
| 38 |
+
"loss": 4.2406,
|
| 39 |
+
"step": 15
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.032,
|
| 43 |
+
"grad_norm": 5.692902088165283,
|
| 44 |
+
"learning_rate": 1.9991650911776697e-05,
|
| 45 |
+
"loss": 4.0798,
|
| 46 |
+
"step": 20
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.032,
|
| 50 |
+
"eval_loss": 4.059751033782959,
|
| 51 |
+
"eval_runtime": 112.8939,
|
| 52 |
+
"eval_samples_per_second": 4.429,
|
| 53 |
+
"eval_steps_per_second": 4.429,
|
| 54 |
+
"step": 20
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.04,
|
| 58 |
+
"grad_norm": 5.954951763153076,
|
| 59 |
+
"learning_rate": 1.9977958222476374e-05,
|
| 60 |
+
"loss": 4.0475,
|
| 61 |
+
"step": 25
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.048,
|
| 65 |
+
"grad_norm": 5.692638874053955,
|
| 66 |
+
"learning_rate": 1.9957756633258264e-05,
|
| 67 |
+
"loss": 3.9848,
|
| 68 |
+
"step": 30
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.048,
|
| 72 |
+
"eval_loss": 3.947784900665283,
|
| 73 |
+
"eval_runtime": 110.4433,
|
| 74 |
+
"eval_samples_per_second": 4.527,
|
| 75 |
+
"eval_steps_per_second": 4.527,
|
| 76 |
+
"step": 30
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.056,
|
| 80 |
+
"grad_norm": 6.3086256980896,
|
| 81 |
+
"learning_rate": 1.9931059322181396e-05,
|
| 82 |
+
"loss": 3.9191,
|
| 83 |
+
"step": 35
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.064,
|
| 87 |
+
"grad_norm": 4.923604965209961,
|
| 88 |
+
"learning_rate": 1.9897883704644982e-05,
|
| 89 |
+
"loss": 3.9104,
|
| 90 |
+
"step": 40
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.064,
|
| 94 |
+
"eval_loss": 3.9222826957702637,
|
| 95 |
+
"eval_runtime": 112.332,
|
| 96 |
+
"eval_samples_per_second": 4.451,
|
| 97 |
+
"eval_steps_per_second": 4.451,
|
| 98 |
+
"step": 40
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.072,
|
| 102 |
+
"grad_norm": 5.413952350616455,
|
| 103 |
+
"learning_rate": 1.9858251422027903e-05,
|
| 104 |
+
"loss": 3.8996,
|
| 105 |
+
"step": 45
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.08,
|
| 109 |
+
"grad_norm": 5.031176567077637,
|
| 110 |
+
"learning_rate": 1.98121883275714e-05,
|
| 111 |
+
"loss": 3.8933,
|
| 112 |
+
"step": 50
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.08,
|
| 116 |
+
"eval_loss": 3.8840794563293457,
|
| 117 |
+
"eval_runtime": 111.9613,
|
| 118 |
+
"eval_samples_per_second": 4.466,
|
| 119 |
+
"eval_steps_per_second": 4.466,
|
| 120 |
+
"step": 50
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.088,
|
| 124 |
+
"grad_norm": 5.418991565704346,
|
| 125 |
+
"learning_rate": 1.9759724469514302e-05,
|
| 126 |
+
"loss": 3.8838,
|
| 127 |
+
"step": 55
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.096,
|
| 131 |
+
"grad_norm": 5.6602864265441895,
|
| 132 |
+
"learning_rate": 1.9700894071491736e-05,
|
| 133 |
+
"loss": 3.8928,
|
| 134 |
+
"step": 60
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.096,
|
| 138 |
+
"eval_loss": 3.859248638153076,
|
| 139 |
+
"eval_runtime": 111.6304,
|
| 140 |
+
"eval_samples_per_second": 4.479,
|
| 141 |
+
"eval_steps_per_second": 4.479,
|
| 142 |
+
"step": 60
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.104,
|
| 146 |
+
"grad_norm": 4.850898265838623,
|
| 147 |
+
"learning_rate": 1.9635735510210053e-05,
|
| 148 |
+
"loss": 3.8717,
|
| 149 |
+
"step": 65
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.112,
|
| 153 |
+
"grad_norm": 4.956837177276611,
|
| 154 |
+
"learning_rate": 1.956429129041269e-05,
|
| 155 |
+
"loss": 3.8501,
|
| 156 |
+
"step": 70
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.112,
|
| 160 |
+
"eval_loss": 3.862858533859253,
|
| 161 |
+
"eval_runtime": 111.0424,
|
| 162 |
+
"eval_samples_per_second": 4.503,
|
| 163 |
+
"eval_steps_per_second": 4.503,
|
| 164 |
+
"step": 70
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.12,
|
| 168 |
+
"grad_norm": 5.260834693908691,
|
| 169 |
+
"learning_rate": 1.94866080171531e-05,
|
| 170 |
+
"loss": 3.8313,
|
| 171 |
+
"step": 75
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.128,
|
| 175 |
+
"grad_norm": 5.144981384277344,
|
| 176 |
+
"learning_rate": 1.940273636539301e-05,
|
| 177 |
+
"loss": 3.8422,
|
| 178 |
+
"step": 80
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.128,
|
| 182 |
+
"eval_loss": 3.8454525470733643,
|
| 183 |
+
"eval_runtime": 112.0635,
|
| 184 |
+
"eval_samples_per_second": 4.462,
|
| 185 |
+
"eval_steps_per_second": 4.462,
|
| 186 |
+
"step": 80
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.136,
|
| 190 |
+
"grad_norm": 5.56984806060791,
|
| 191 |
+
"learning_rate": 1.9312731046945694e-05,
|
| 192 |
+
"loss": 3.836,
|
| 193 |
+
"step": 85
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.144,
|
| 197 |
+
"grad_norm": 4.828178882598877,
|
| 198 |
+
"learning_rate": 1.9216650774785975e-05,
|
| 199 |
+
"loss": 3.8071,
|
| 200 |
+
"step": 90
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.144,
|
| 204 |
+
"eval_loss": 3.831526041030884,
|
| 205 |
+
"eval_runtime": 111.1507,
|
| 206 |
+
"eval_samples_per_second": 4.498,
|
| 207 |
+
"eval_steps_per_second": 4.498,
|
| 208 |
+
"step": 90
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.152,
|
| 212 |
+
"grad_norm": 5.023971080780029,
|
| 213 |
+
"learning_rate": 1.911455822475009e-05,
|
| 214 |
+
"loss": 3.8,
|
| 215 |
+
"step": 95
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.16,
|
| 219 |
+
"grad_norm": 4.5282793045043945,
|
| 220 |
+
"learning_rate": 1.9006519994650516e-05,
|
| 221 |
+
"loss": 3.7826,
|
| 222 |
+
"step": 100
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.16,
|
| 226 |
+
"eval_loss": 3.8195042610168457,
|
| 227 |
+
"eval_runtime": 109.8545,
|
| 228 |
+
"eval_samples_per_second": 4.551,
|
| 229 |
+
"eval_steps_per_second": 4.551,
|
| 230 |
+
"step": 100
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.168,
|
| 234 |
+
"grad_norm": 5.027553081512451,
|
| 235 |
+
"learning_rate": 1.8892606560832335e-05,
|
| 236 |
+
"loss": 3.7953,
|
| 237 |
+
"step": 105
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.176,
|
| 241 |
+
"grad_norm": 4.897686004638672,
|
| 242 |
+
"learning_rate": 1.8772892232199594e-05,
|
| 243 |
+
"loss": 3.8194,
|
| 244 |
+
"step": 110
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.176,
|
| 248 |
+
"eval_loss": 3.784461498260498,
|
| 249 |
+
"eval_runtime": 112.129,
|
| 250 |
+
"eval_samples_per_second": 4.459,
|
| 251 |
+
"eval_steps_per_second": 4.459,
|
| 252 |
+
"step": 110
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.184,
|
| 256 |
+
"grad_norm": 4.60252046585083,
|
| 257 |
+
"learning_rate": 1.8647455101741492e-05,
|
| 258 |
+
"loss": 3.774,
|
| 259 |
+
"step": 115
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.192,
|
| 263 |
+
"grad_norm": 4.88078498840332,
|
| 264 |
+
"learning_rate": 1.8516376995590185e-05,
|
| 265 |
+
"loss": 3.7932,
|
| 266 |
+
"step": 120
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.192,
|
| 270 |
+
"eval_loss": 3.8124334812164307,
|
| 271 |
+
"eval_runtime": 111.9015,
|
| 272 |
+
"eval_samples_per_second": 4.468,
|
| 273 |
+
"eval_steps_per_second": 4.468,
|
| 274 |
+
"step": 120
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.2,
|
| 278 |
+
"grad_norm": 5.119283676147461,
|
| 279 |
+
"learning_rate": 1.837974341964331e-05,
|
| 280 |
+
"loss": 3.7665,
|
| 281 |
+
"step": 125
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.208,
|
| 285 |
+
"grad_norm": 4.381007671356201,
|
| 286 |
+
"learning_rate": 1.8237643503786097e-05,
|
| 287 |
+
"loss": 3.7776,
|
| 288 |
+
"step": 130
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.208,
|
| 292 |
+
"eval_loss": 3.7836484909057617,
|
| 293 |
+
"eval_runtime": 111.6098,
|
| 294 |
+
"eval_samples_per_second": 4.48,
|
| 295 |
+
"eval_steps_per_second": 4.48,
|
| 296 |
+
"step": 130
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.216,
|
| 300 |
+
"grad_norm": 4.681989669799805,
|
| 301 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 302 |
+
"loss": 3.7758,
|
| 303 |
+
"step": 135
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.224,
|
| 307 |
+
"grad_norm": 4.945594310760498,
|
| 308 |
+
"learning_rate": 1.7937418940642076e-05,
|
| 309 |
+
"loss": 3.7704,
|
| 310 |
+
"step": 140
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.224,
|
| 314 |
+
"eval_loss": 3.781540632247925,
|
| 315 |
+
"eval_runtime": 111.7063,
|
| 316 |
+
"eval_samples_per_second": 4.476,
|
| 317 |
+
"eval_steps_per_second": 4.476,
|
| 318 |
+
"step": 140
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.232,
|
| 322 |
+
"grad_norm": 4.896081924438477,
|
| 323 |
+
"learning_rate": 1.7779490138195544e-05,
|
| 324 |
+
"loss": 3.7896,
|
| 325 |
+
"step": 145
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.24,
|
| 329 |
+
"grad_norm": 4.702622890472412,
|
| 330 |
+
"learning_rate": 1.7616486557764187e-05,
|
| 331 |
+
"loss": 3.7802,
|
| 332 |
+
"step": 150
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.24,
|
| 336 |
+
"eval_loss": 3.7781624794006348,
|
| 337 |
+
"eval_runtime": 112.1095,
|
| 338 |
+
"eval_samples_per_second": 4.46,
|
| 339 |
+
"eval_steps_per_second": 4.46,
|
| 340 |
+
"step": 150
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.248,
|
| 344 |
+
"grad_norm": 4.960013389587402,
|
| 345 |
+
"learning_rate": 1.7448514531121265e-05,
|
| 346 |
+
"loss": 3.7478,
|
| 347 |
+
"step": 155
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.256,
|
| 351 |
+
"grad_norm": 4.85302734375,
|
| 352 |
+
"learning_rate": 1.72756836310958e-05,
|
| 353 |
+
"loss": 3.7776,
|
| 354 |
+
"step": 160
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.256,
|
| 358 |
+
"eval_loss": 3.767767906188965,
|
| 359 |
+
"eval_runtime": 111.5609,
|
| 360 |
+
"eval_samples_per_second": 4.482,
|
| 361 |
+
"eval_steps_per_second": 4.482,
|
| 362 |
+
"step": 160
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.264,
|
| 366 |
+
"grad_norm": 4.286273002624512,
|
| 367 |
+
"learning_rate": 1.7098106600095204e-05,
|
| 368 |
+
"loss": 3.7231,
|
| 369 |
+
"step": 165
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.272,
|
| 373 |
+
"grad_norm": 5.215882778167725,
|
| 374 |
+
"learning_rate": 1.6952705502317577e-05,
|
| 375 |
+
"loss": 3.777,
|
| 376 |
+
"step": 170
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.272,
|
| 380 |
+
"eval_loss": 3.758565902709961,
|
| 381 |
+
"eval_runtime": 112.4381,
|
| 382 |
+
"eval_samples_per_second": 4.447,
|
| 383 |
+
"eval_steps_per_second": 4.447,
|
| 384 |
+
"step": 170
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.28,
|
| 388 |
+
"grad_norm": 4.810149669647217,
|
| 389 |
+
"learning_rate": 1.6766879380776983e-05,
|
| 390 |
+
"loss": 3.7346,
|
| 391 |
+
"step": 175
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.288,
|
| 395 |
+
"grad_norm": 4.446249961853027,
|
| 396 |
+
"learning_rate": 1.6576639035437975e-05,
|
| 397 |
+
"loss": 3.7554,
|
| 398 |
+
"step": 180
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.288,
|
| 402 |
+
"eval_loss": 3.7531182765960693,
|
| 403 |
+
"eval_runtime": 111.8589,
|
| 404 |
+
"eval_samples_per_second": 4.47,
|
| 405 |
+
"eval_steps_per_second": 4.47,
|
| 406 |
+
"step": 180
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.296,
|
| 410 |
+
"grad_norm": 4.627930164337158,
|
| 411 |
+
"learning_rate": 1.6382108565373783e-05,
|
| 412 |
+
"loss": 3.7298,
|
| 413 |
+
"step": 185
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.304,
|
| 417 |
+
"grad_norm": 4.859839916229248,
|
| 418 |
+
"learning_rate": 1.6183414868225434e-05,
|
| 419 |
+
"loss": 3.7494,
|
| 420 |
+
"step": 190
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.304,
|
| 424 |
+
"eval_loss": 3.754574775695801,
|
| 425 |
+
"eval_runtime": 111.1646,
|
| 426 |
+
"eval_samples_per_second": 4.498,
|
| 427 |
+
"eval_steps_per_second": 4.498,
|
| 428 |
+
"step": 190
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.312,
|
| 432 |
+
"grad_norm": 4.594749927520752,
|
| 433 |
+
"learning_rate": 1.5980687557422854e-05,
|
| 434 |
+
"loss": 3.7233,
|
| 435 |
+
"step": 195
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.32,
|
| 439 |
+
"grad_norm": 4.1947479248046875,
|
| 440 |
+
"learning_rate": 1.5774058877634473e-05,
|
| 441 |
+
"loss": 3.7535,
|
| 442 |
+
"step": 200
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.32,
|
| 446 |
+
"eval_loss": 3.740143299102783,
|
| 447 |
+
"eval_runtime": 110.7913,
|
| 448 |
+
"eval_samples_per_second": 4.513,
|
| 449 |
+
"eval_steps_per_second": 4.513,
|
| 450 |
+
"step": 200
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.328,
|
| 454 |
+
"grad_norm": 5.630258560180664,
|
| 455 |
+
"learning_rate": 1.55636636185003e-05,
|
| 456 |
+
"loss": 3.7192,
|
| 457 |
+
"step": 205
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.336,
|
| 461 |
+
"grad_norm": 5.389300346374512,
|
| 462 |
+
"learning_rate": 1.5349639026704916e-05,
|
| 463 |
+
"loss": 3.7408,
|
| 464 |
+
"step": 210
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.336,
|
| 468 |
+
"eval_loss": 3.7239396572113037,
|
| 469 |
+
"eval_runtime": 113.2108,
|
| 470 |
+
"eval_samples_per_second": 4.417,
|
| 471 |
+
"eval_steps_per_second": 4.417,
|
| 472 |
+
"step": 210
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.344,
|
| 476 |
+
"grad_norm": 4.697848796844482,
|
| 477 |
+
"learning_rate": 1.5132124716447627e-05,
|
| 478 |
+
"loss": 3.7209,
|
| 479 |
+
"step": 215
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.352,
|
| 483 |
+
"grad_norm": 4.92866325378418,
|
| 484 |
+
"learning_rate": 1.4911262578368233e-05,
|
| 485 |
+
"loss": 3.7015,
|
| 486 |
+
"step": 220
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.352,
|
| 490 |
+
"eval_loss": 3.724281072616577,
|
| 491 |
+
"eval_runtime": 112.1344,
|
| 492 |
+
"eval_samples_per_second": 4.459,
|
| 493 |
+
"eval_steps_per_second": 4.459,
|
| 494 |
+
"step": 220
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.36,
|
| 498 |
+
"grad_norm": 5.7383036613464355,
|
| 499 |
+
"learning_rate": 1.4687196686987832e-05,
|
| 500 |
+
"loss": 3.7033,
|
| 501 |
+
"step": 225
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.368,
|
| 505 |
+
"grad_norm": 5.267530918121338,
|
| 506 |
+
"learning_rate": 1.4460073206724984e-05,
|
| 507 |
+
"loss": 3.7526,
|
| 508 |
+
"step": 230
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.368,
|
| 512 |
+
"eval_loss": 3.7359769344329834,
|
| 513 |
+
"eval_runtime": 111.4037,
|
| 514 |
+
"eval_samples_per_second": 4.488,
|
| 515 |
+
"eval_steps_per_second": 4.488,
|
| 516 |
+
"step": 230
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.376,
|
| 520 |
+
"grad_norm": 4.392041206359863,
|
| 521 |
+
"learning_rate": 1.4230040296548588e-05,
|
| 522 |
+
"loss": 3.7569,
|
| 523 |
+
"step": 235
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.384,
|
| 527 |
+
"grad_norm": 4.574108600616455,
|
| 528 |
+
"learning_rate": 1.3997248013329681e-05,
|
| 529 |
+
"loss": 3.7129,
|
| 530 |
+
"step": 240
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.384,
|
| 534 |
+
"eval_loss": 3.7244505882263184,
|
| 535 |
+
"eval_runtime": 111.6979,
|
| 536 |
+
"eval_samples_per_second": 4.476,
|
| 537 |
+
"eval_steps_per_second": 4.476,
|
| 538 |
+
"step": 240
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.392,
|
| 542 |
+
"grad_norm": 4.590445041656494,
|
| 543 |
+
"learning_rate": 1.3761848213955136e-05,
|
| 544 |
+
"loss": 3.6811,
|
| 545 |
+
"step": 245
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.4,
|
| 549 |
+
"grad_norm": 4.465660095214844,
|
| 550 |
+
"learning_rate": 1.352399445626722e-05,
|
| 551 |
+
"loss": 3.7496,
|
| 552 |
+
"step": 250
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.4,
|
| 556 |
+
"eval_loss": 3.721489191055298,
|
| 557 |
+
"eval_runtime": 112.1994,
|
| 558 |
+
"eval_samples_per_second": 4.456,
|
| 559 |
+
"eval_steps_per_second": 4.456,
|
| 560 |
+
"step": 250
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.408,
|
| 564 |
+
"grad_norm": 4.747437953948975,
|
| 565 |
+
"learning_rate": 1.328384189889351e-05,
|
| 566 |
+
"loss": 3.7119,
|
| 567 |
+
"step": 255
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 0.416,
|
| 571 |
+
"grad_norm": 5.685433387756348,
|
| 572 |
+
"learning_rate": 1.3041547200032618e-05,
|
| 573 |
+
"loss": 3.6913,
|
| 574 |
+
"step": 260
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.416,
|
| 578 |
+
"eval_loss": 3.7043662071228027,
|
| 579 |
+
"eval_runtime": 111.183,
|
| 580 |
+
"eval_samples_per_second": 4.497,
|
| 581 |
+
"eval_steps_per_second": 4.497,
|
| 582 |
+
"step": 260
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 0.424,
|
| 586 |
+
"grad_norm": 4.718033313751221,
|
| 587 |
+
"learning_rate": 1.2797268415261681e-05,
|
| 588 |
+
"loss": 3.6865,
|
| 589 |
+
"step": 265
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.432,
|
| 593 |
+
"grad_norm": 4.735610485076904,
|
| 594 |
+
"learning_rate": 1.2551164894432305e-05,
|
| 595 |
+
"loss": 3.6844,
|
| 596 |
+
"step": 270
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.432,
|
| 600 |
+
"eval_loss": 3.7100021839141846,
|
| 601 |
+
"eval_runtime": 111.4098,
|
| 602 |
+
"eval_samples_per_second": 4.488,
|
| 603 |
+
"eval_steps_per_second": 4.488,
|
| 604 |
+
"step": 270
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.44,
|
| 608 |
+
"grad_norm": 4.772237777709961,
|
| 609 |
+
"learning_rate": 1.2303397177722234e-05,
|
| 610 |
+
"loss": 3.6997,
|
| 611 |
+
"step": 275
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.448,
|
| 615 |
+
"grad_norm": 5.533233642578125,
|
| 616 |
+
"learning_rate": 1.2054126890910499e-05,
|
| 617 |
+
"loss": 3.7172,
|
| 618 |
+
"step": 280
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.448,
|
| 622 |
+
"eval_loss": 3.7194995880126953,
|
| 623 |
+
"eval_runtime": 112.4891,
|
| 624 |
+
"eval_samples_per_second": 4.445,
|
| 625 |
+
"eval_steps_per_second": 4.445,
|
| 626 |
+
"step": 280
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.456,
|
| 630 |
+
"grad_norm": 4.787185192108154,
|
| 631 |
+
"learning_rate": 1.1803516639944452e-05,
|
| 632 |
+
"loss": 3.7097,
|
| 633 |
+
"step": 285
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.464,
|
| 637 |
+
"grad_norm": 4.525595664978027,
|
| 638 |
+
"learning_rate": 1.1551729904867353e-05,
|
| 639 |
+
"loss": 3.739,
|
| 640 |
+
"step": 290
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.464,
|
| 644 |
+
"eval_loss": 3.70326828956604,
|
| 645 |
+
"eval_runtime": 111.8761,
|
| 646 |
+
"eval_samples_per_second": 4.469,
|
| 647 |
+
"eval_steps_per_second": 4.469,
|
| 648 |
+
"step": 290
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.472,
|
| 652 |
+
"grad_norm": 4.284120082855225,
|
| 653 |
+
"learning_rate": 1.1298930933175805e-05,
|
| 654 |
+
"loss": 3.684,
|
| 655 |
+
"step": 295
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.48,
|
| 659 |
+
"grad_norm": 5.107572555541992,
|
| 660 |
+
"learning_rate": 1.1045284632676535e-05,
|
| 661 |
+
"loss": 3.6814,
|
| 662 |
+
"step": 300
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.48,
|
| 666 |
+
"eval_loss": 3.701869249343872,
|
| 667 |
+
"eval_runtime": 110.8874,
|
| 668 |
+
"eval_samples_per_second": 4.509,
|
| 669 |
+
"eval_steps_per_second": 4.509,
|
| 670 |
+
"step": 300
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.488,
|
| 674 |
+
"grad_norm": 4.888582706451416,
|
| 675 |
+
"learning_rate": 1.0790956463912439e-05,
|
| 676 |
+
"loss": 3.6695,
|
| 677 |
+
"step": 305
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.496,
|
| 681 |
+
"grad_norm": 5.606681823730469,
|
| 682 |
+
"learning_rate": 1.0536112332228057e-05,
|
| 683 |
+
"loss": 3.6809,
|
| 684 |
+
"step": 310
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.496,
|
| 688 |
+
"eval_loss": 3.6872060298919678,
|
| 689 |
+
"eval_runtime": 112.8373,
|
| 690 |
+
"eval_samples_per_second": 4.431,
|
| 691 |
+
"eval_steps_per_second": 4.431,
|
| 692 |
+
"step": 310
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.504,
|
| 696 |
+
"grad_norm": 4.867734909057617,
|
| 697 |
+
"learning_rate": 1.0280918479544914e-05,
|
| 698 |
+
"loss": 3.707,
|
| 699 |
+
"step": 315
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.512,
|
| 703 |
+
"grad_norm": 7.736087322235107,
|
| 704 |
+
"learning_rate": 1.0025541375917274e-05,
|
| 705 |
+
"loss": 3.7031,
|
| 706 |
+
"step": 320
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.512,
|
| 710 |
+
"eval_loss": 3.6830782890319824,
|
| 711 |
+
"eval_runtime": 110.3497,
|
| 712 |
+
"eval_samples_per_second": 4.531,
|
| 713 |
+
"eval_steps_per_second": 4.531,
|
| 714 |
+
"step": 320
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.52,
|
| 718 |
+
"grad_norm": 4.444705486297607,
|
| 719 |
+
"learning_rate": 9.770147610939098e-06,
|
| 720 |
+
"loss": 3.7152,
|
| 721 |
+
"step": 325
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.528,
|
| 725 |
+
"grad_norm": 15.434615135192871,
|
| 726 |
+
"learning_rate": 9.514903785072998e-06,
|
| 727 |
+
"loss": 3.7322,
|
| 728 |
+
"step": 330
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.528,
|
| 732 |
+
"eval_loss": 3.698810577392578,
|
| 733 |
+
"eval_runtime": 111.3388,
|
| 734 |
+
"eval_samples_per_second": 4.491,
|
| 735 |
+
"eval_steps_per_second": 4.491,
|
| 736 |
+
"step": 330
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.536,
|
| 740 |
+
"grad_norm": 5.347226142883301,
|
| 741 |
+
"learning_rate": 9.259976400972147e-06,
|
| 742 |
+
"loss": 3.6663,
|
| 743 |
+
"step": 335
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.544,
|
| 747 |
+
"grad_norm": 4.670411109924316,
|
| 748 |
+
"learning_rate": 9.005531754865929e-06,
|
| 749 |
+
"loss": 3.6813,
|
| 750 |
+
"step": 340
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.544,
|
| 754 |
+
"eval_loss": 3.7001984119415283,
|
| 755 |
+
"eval_runtime": 113.107,
|
| 756 |
+
"eval_samples_per_second": 4.421,
|
| 757 |
+
"eval_steps_per_second": 4.421,
|
| 758 |
+
"step": 340
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.552,
|
| 762 |
+
"grad_norm": 5.2611589431762695,
|
| 763 |
+
"learning_rate": 8.751735828080308e-06,
|
| 764 |
+
"loss": 3.6737,
|
| 765 |
+
"step": 345
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.56,
|
| 769 |
+
"grad_norm": 4.978248596191406,
|
| 770 |
+
"learning_rate": 8.498754178763564e-06,
|
| 771 |
+
"loss": 3.6661,
|
| 772 |
+
"step": 350
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.56,
|
| 776 |
+
"eval_loss": 3.6846611499786377,
|
| 777 |
+
"eval_runtime": 111.743,
|
| 778 |
+
"eval_samples_per_second": 4.475,
|
| 779 |
+
"eval_steps_per_second": 4.475,
|
| 780 |
+
"step": 350
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.568,
|
| 784 |
+
"grad_norm": 5.137356758117676,
|
| 785 |
+
"learning_rate": 8.246751833888122e-06,
|
| 786 |
+
"loss": 3.6944,
|
| 787 |
+
"step": 355
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.576,
|
| 791 |
+
"grad_norm": 5.919885635375977,
|
| 792 |
+
"learning_rate": 7.995893181598866e-06,
|
| 793 |
+
"loss": 3.7133,
|
| 794 |
+
"step": 360
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.576,
|
| 798 |
+
"eval_loss": 3.682840585708618,
|
| 799 |
+
"eval_runtime": 110.2256,
|
| 800 |
+
"eval_samples_per_second": 4.536,
|
| 801 |
+
"eval_steps_per_second": 4.536,
|
| 802 |
+
"step": 360
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.584,
|
| 806 |
+
"grad_norm": 4.792472839355469,
|
| 807 |
+
"learning_rate": 7.746341863978151e-06,
|
| 808 |
+
"loss": 3.6581,
|
| 809 |
+
"step": 365
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.592,
|
| 813 |
+
"grad_norm": 4.619278430938721,
|
| 814 |
+
"learning_rate": 7.4982606702975505e-06,
|
| 815 |
+
"loss": 3.7072,
|
| 816 |
+
"step": 370
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.592,
|
| 820 |
+
"eval_loss": 3.6833138465881348,
|
| 821 |
+
"eval_runtime": 112.4846,
|
| 822 |
+
"eval_samples_per_second": 4.445,
|
| 823 |
+
"eval_steps_per_second": 4.445,
|
| 824 |
+
"step": 370
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.6,
|
| 828 |
+
"grad_norm": 5.981200218200684,
|
| 829 |
+
"learning_rate": 7.251811430825846e-06,
|
| 830 |
+
"loss": 3.6818,
|
| 831 |
+
"step": 375
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.608,
|
| 835 |
+
"grad_norm": 5.758677005767822,
|
| 836 |
+
"learning_rate": 7.007154911262678e-06,
|
| 837 |
+
"loss": 3.665,
|
| 838 |
+
"step": 380
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.608,
|
| 842 |
+
"eval_loss": 3.6854116916656494,
|
| 843 |
+
"eval_runtime": 112.1801,
|
| 844 |
+
"eval_samples_per_second": 4.457,
|
| 845 |
+
"eval_steps_per_second": 4.457,
|
| 846 |
+
"step": 380
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 0.616,
|
| 850 |
+
"grad_norm": 4.582080841064453,
|
| 851 |
+
"learning_rate": 6.764450707866577e-06,
|
| 852 |
+
"loss": 3.667,
|
| 853 |
+
"step": 385
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.624,
|
| 857 |
+
"grad_norm": 5.349440574645996,
|
| 858 |
+
"learning_rate": 6.523857143345884e-06,
|
| 859 |
+
"loss": 3.6779,
|
| 860 |
+
"step": 390
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.624,
|
| 864 |
+
"eval_loss": 3.67708158493042,
|
| 865 |
+
"eval_runtime": 111.3211,
|
| 866 |
+
"eval_samples_per_second": 4.492,
|
| 867 |
+
"eval_steps_per_second": 4.492,
|
| 868 |
+
"step": 390
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 0.632,
|
| 872 |
+
"grad_norm": 4.663772106170654,
|
| 873 |
+
"learning_rate": 6.2855311635804495e-06,
|
| 874 |
+
"loss": 3.6561,
|
| 875 |
+
"step": 395
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 0.64,
|
| 879 |
+
"grad_norm": 9.236244201660156,
|
| 880 |
+
"learning_rate": 6.049628235241459e-06,
|
| 881 |
+
"loss": 3.6556,
|
| 882 |
+
"step": 400
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 0.64,
|
| 886 |
+
"eval_loss": 3.6733860969543457,
|
| 887 |
+
"eval_runtime": 111.5456,
|
| 888 |
+
"eval_samples_per_second": 4.482,
|
| 889 |
+
"eval_steps_per_second": 4.482,
|
| 890 |
+
"step": 400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.648,
|
| 894 |
+
"grad_norm": 5.29210090637207,
|
| 895 |
+
"learning_rate": 5.816302244376206e-06,
|
| 896 |
+
"loss": 3.6791,
|
| 897 |
+
"step": 405
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.656,
|
| 901 |
+
"grad_norm": 4.581356048583984,
|
| 902 |
+
"learning_rate": 5.585705396023931e-06,
|
| 903 |
+
"loss": 3.6803,
|
| 904 |
+
"step": 410
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.656,
|
| 908 |
+
"eval_loss": 3.67871356010437,
|
| 909 |
+
"eval_runtime": 110.9923,
|
| 910 |
+
"eval_samples_per_second": 4.505,
|
| 911 |
+
"eval_steps_per_second": 4.505,
|
| 912 |
+
"step": 410
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.664,
|
| 916 |
+
"grad_norm": 5.390838623046875,
|
| 917 |
+
"learning_rate": 5.357988114928221e-06,
|
| 918 |
+
"loss": 3.6861,
|
| 919 |
+
"step": 415
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.672,
|
| 923 |
+
"grad_norm": 4.902568340301514,
|
| 924 |
+
"learning_rate": 5.133298947410812e-06,
|
| 925 |
+
"loss": 3.6451,
|
| 926 |
+
"step": 420
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.672,
|
| 930 |
+
"eval_loss": 3.666407585144043,
|
| 931 |
+
"eval_runtime": 112.3761,
|
| 932 |
+
"eval_samples_per_second": 4.449,
|
| 933 |
+
"eval_steps_per_second": 4.449,
|
| 934 |
+
"step": 420
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.68,
|
| 938 |
+
"grad_norm": 4.62513542175293,
|
| 939 |
+
"learning_rate": 4.911784464470624e-06,
|
| 940 |
+
"loss": 3.69,
|
| 941 |
+
"step": 425
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.688,
|
| 945 |
+
"grad_norm": 4.63794469833374,
|
| 946 |
+
"learning_rate": 4.693589166171466e-06,
|
| 947 |
+
"loss": 3.6554,
|
| 948 |
+
"step": 430
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.688,
|
| 952 |
+
"eval_loss": 3.6700761318206787,
|
| 953 |
+
"eval_runtime": 110.8149,
|
| 954 |
+
"eval_samples_per_second": 4.512,
|
| 955 |
+
"eval_steps_per_second": 4.512,
|
| 956 |
+
"step": 430
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.696,
|
| 960 |
+
"grad_norm": 5.994736671447754,
|
| 961 |
+
"learning_rate": 4.4788553873806054e-06,
|
| 962 |
+
"loss": 3.6735,
|
| 963 |
+
"step": 435
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.704,
|
| 967 |
+
"grad_norm": 4.715162754058838,
|
| 968 |
+
"learning_rate": 4.267723204919764e-06,
|
| 969 |
+
"loss": 3.6115,
|
| 970 |
+
"step": 440
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.704,
|
| 974 |
+
"eval_loss": 3.660256862640381,
|
| 975 |
+
"eval_runtime": 112.647,
|
| 976 |
+
"eval_samples_per_second": 4.439,
|
| 977 |
+
"eval_steps_per_second": 4.439,
|
| 978 |
+
"step": 440
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.712,
|
| 982 |
+
"grad_norm": 8.556255340576172,
|
| 983 |
+
"learning_rate": 4.060330346189125e-06,
|
| 984 |
+
"loss": 3.6289,
|
| 985 |
+
"step": 445
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.72,
|
| 989 |
+
"grad_norm": 4.761578559875488,
|
| 990 |
+
"learning_rate": 3.8568120993238936e-06,
|
| 991 |
+
"loss": 3.6678,
|
| 992 |
+
"step": 450
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.72,
|
| 996 |
+
"eval_loss": 3.67647385597229,
|
| 997 |
+
"eval_runtime": 110.1504,
|
| 998 |
+
"eval_samples_per_second": 4.539,
|
| 999 |
+
"eval_steps_per_second": 4.539,
|
| 1000 |
+
"step": 450
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.728,
|
| 1004 |
+
"grad_norm": 4.757038593292236,
|
| 1005 |
+
"learning_rate": 3.657301224942098e-06,
|
| 1006 |
+
"loss": 3.7008,
|
| 1007 |
+
"step": 455
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.736,
|
| 1011 |
+
"grad_norm": 4.461075305938721,
|
| 1012 |
+
"learning_rate": 3.4619278695411495e-06,
|
| 1013 |
+
"loss": 3.651,
|
| 1014 |
+
"step": 460
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.736,
|
| 1018 |
+
"eval_loss": 3.6577467918395996,
|
| 1019 |
+
"eval_runtime": 110.5987,
|
| 1020 |
+
"eval_samples_per_second": 4.521,
|
| 1021 |
+
"eval_steps_per_second": 4.521,
|
| 1022 |
+
"step": 460
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.744,
|
| 1026 |
+
"grad_norm": 3.8970530033111572,
|
| 1027 |
+
"learning_rate": 3.2708194805996252e-06,
|
| 1028 |
+
"loss": 3.6794,
|
| 1029 |
+
"step": 465
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.752,
|
| 1033 |
+
"grad_norm": 4.632080554962158,
|
| 1034 |
+
"learning_rate": 3.0841007234397655e-06,
|
| 1035 |
+
"loss": 3.6475,
|
| 1036 |
+
"step": 470
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.752,
|
| 1040 |
+
"eval_loss": 3.661729574203491,
|
| 1041 |
+
"eval_runtime": 112.6803,
|
| 1042 |
+
"eval_samples_per_second": 4.437,
|
| 1043 |
+
"eval_steps_per_second": 4.437,
|
| 1044 |
+
"step": 470
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.76,
|
| 1048 |
+
"grad_norm": 5.098201751708984,
|
| 1049 |
+
"learning_rate": 2.901893399904797e-06,
|
| 1050 |
+
"loss": 3.6647,
|
| 1051 |
+
"step": 475
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.768,
|
| 1055 |
+
"grad_norm": 4.249239921569824,
|
| 1056 |
+
"learning_rate": 2.724316368904201e-06,
|
| 1057 |
+
"loss": 3.694,
|
| 1058 |
+
"step": 480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.768,
|
| 1062 |
+
"eval_loss": 3.652348041534424,
|
| 1063 |
+
"eval_runtime": 111.7392,
|
| 1064 |
+
"eval_samples_per_second": 4.475,
|
| 1065 |
+
"eval_steps_per_second": 4.475,
|
| 1066 |
+
"step": 480
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.776,
|
| 1070 |
+
"grad_norm": 4.952099800109863,
|
| 1071 |
+
"learning_rate": 2.5514854688787406e-06,
|
| 1072 |
+
"loss": 3.6771,
|
| 1073 |
+
"step": 485
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.784,
|
| 1077 |
+
"grad_norm": 4.995110034942627,
|
| 1078 |
+
"learning_rate": 2.383513442235812e-06,
|
| 1079 |
+
"loss": 3.6666,
|
| 1080 |
+
"step": 490
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.784,
|
| 1084 |
+
"eval_loss": 3.653099298477173,
|
| 1085 |
+
"eval_runtime": 111.5607,
|
| 1086 |
+
"eval_samples_per_second": 4.482,
|
| 1087 |
+
"eval_steps_per_second": 4.482,
|
| 1088 |
+
"step": 490
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.792,
|
| 1092 |
+
"grad_norm": 5.493462085723877,
|
| 1093 |
+
"learning_rate": 2.2205098618044584e-06,
|
| 1094 |
+
"loss": 3.6716,
|
| 1095 |
+
"step": 495
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.8,
|
| 1099 |
+
"grad_norm": 4.578746795654297,
|
| 1100 |
+
"learning_rate": 2.0625810593579286e-06,
|
| 1101 |
+
"loss": 3.64,
|
| 1102 |
+
"step": 500
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.8,
|
| 1106 |
+
"eval_loss": 3.6576831340789795,
|
| 1107 |
+
"eval_runtime": 109.6662,
|
| 1108 |
+
"eval_samples_per_second": 4.559,
|
| 1109 |
+
"eval_steps_per_second": 4.559,
|
| 1110 |
+
"step": 500
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.808,
|
| 1114 |
+
"grad_norm": 4.769631385803223,
|
| 1115 |
+
"learning_rate": 1.9098300562505266e-06,
|
| 1116 |
+
"loss": 3.6298,
|
| 1117 |
+
"step": 505
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.816,
|
| 1121 |
+
"grad_norm": 5.270910263061523,
|
| 1122 |
+
"learning_rate": 1.7623564962139061e-06,
|
| 1123 |
+
"loss": 3.6375,
|
| 1124 |
+
"step": 510
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.816,
|
| 1128 |
+
"eval_loss": 3.658618211746216,
|
| 1129 |
+
"eval_runtime": 112.7334,
|
| 1130 |
+
"eval_samples_per_second": 4.435,
|
| 1131 |
+
"eval_steps_per_second": 4.435,
|
| 1132 |
+
"step": 510
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 0.824,
|
| 1136 |
+
"grad_norm": 4.440089225769043,
|
| 1137 |
+
"learning_rate": 1.6202565803566917e-06,
|
| 1138 |
+
"loss": 3.6672,
|
| 1139 |
+
"step": 515
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 0.832,
|
| 1143 |
+
"grad_norm": 4.2689337730407715,
|
| 1144 |
+
"learning_rate": 1.4836230044098164e-06,
|
| 1145 |
+
"loss": 3.6544,
|
| 1146 |
+
"step": 520
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 0.832,
|
| 1150 |
+
"eval_loss": 3.6577441692352295,
|
| 1151 |
+
"eval_runtime": 111.1537,
|
| 1152 |
+
"eval_samples_per_second": 4.498,
|
| 1153 |
+
"eval_steps_per_second": 4.498,
|
| 1154 |
+
"step": 520
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.84,
|
| 1158 |
+
"grad_norm": 7.282012939453125,
|
| 1159 |
+
"learning_rate": 1.352544898258511e-06,
|
| 1160 |
+
"loss": 3.6416,
|
| 1161 |
+
"step": 525
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.848,
|
| 1165 |
+
"grad_norm": 5.834847927093506,
|
| 1166 |
+
"learning_rate": 1.2271077678004084e-06,
|
| 1167 |
+
"loss": 3.653,
|
| 1168 |
+
"step": 530
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.848,
|
| 1172 |
+
"eval_loss": 3.6557505130767822,
|
| 1173 |
+
"eval_runtime": 111.4698,
|
| 1174 |
+
"eval_samples_per_second": 4.486,
|
| 1175 |
+
"eval_steps_per_second": 4.486,
|
| 1176 |
+
"step": 530
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 0.856,
|
| 1180 |
+
"grad_norm": 4.881261825561523,
|
| 1181 |
+
"learning_rate": 1.1073934391676666e-06,
|
| 1182 |
+
"loss": 3.6859,
|
| 1183 |
+
"step": 535
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 0.864,
|
| 1187 |
+
"grad_norm": 4.997443675994873,
|
| 1188 |
+
"learning_rate": 9.934800053494886e-07,
|
| 1189 |
+
"loss": 3.6678,
|
| 1190 |
+
"step": 540
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.864,
|
| 1194 |
+
"eval_loss": 3.6547605991363525,
|
| 1195 |
+
"eval_runtime": 112.6768,
|
| 1196 |
+
"eval_samples_per_second": 4.437,
|
| 1197 |
+
"eval_steps_per_second": 4.437,
|
| 1198 |
+
"step": 540
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.872,
|
| 1202 |
+
"grad_norm": 5.601071834564209,
|
| 1203 |
+
"learning_rate": 8.854417752499112e-07,
|
| 1204 |
+
"loss": 3.6535,
|
| 1205 |
+
"step": 545
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.88,
|
| 1209 |
+
"grad_norm": 5.277810573577881,
|
| 1210 |
+
"learning_rate": 7.833492252140284e-07,
|
| 1211 |
+
"loss": 3.6455,
|
| 1212 |
+
"step": 550
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.88,
|
| 1216 |
+
"eval_loss": 3.6565444469451904,
|
| 1217 |
+
"eval_runtime": 111.5954,
|
| 1218 |
+
"eval_samples_per_second": 4.48,
|
| 1219 |
+
"eval_steps_per_second": 4.48,
|
| 1220 |
+
"step": 550
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.888,
|
| 1224 |
+
"grad_norm": 4.55835485458374,
|
| 1225 |
+
"learning_rate": 6.872689530543087e-07,
|
| 1226 |
+
"loss": 3.6681,
|
| 1227 |
+
"step": 555
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.896,
|
| 1231 |
+
"grad_norm": 4.985684871673584,
|
| 1232 |
+
"learning_rate": 5.972636346069949e-07,
|
| 1233 |
+
"loss": 3.6437,
|
| 1234 |
+
"step": 560
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.896,
|
| 1238 |
+
"eval_loss": 3.656482696533203,
|
| 1239 |
+
"eval_runtime": 111.9622,
|
| 1240 |
+
"eval_samples_per_second": 4.466,
|
| 1241 |
+
"eval_steps_per_second": 4.466,
|
| 1242 |
+
"step": 560
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.904,
|
| 1246 |
+
"grad_norm": 5.561159133911133,
|
| 1247 |
+
"learning_rate": 5.133919828468992e-07,
|
| 1248 |
+
"loss": 3.6635,
|
| 1249 |
+
"step": 565
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.912,
|
| 1253 |
+
"grad_norm": 5.3784003257751465,
|
| 1254 |
+
"learning_rate": 4.357087095873136e-07,
|
| 1255 |
+
"loss": 3.6519,
|
| 1256 |
+
"step": 570
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.912,
|
| 1260 |
+
"eval_loss": 3.656426429748535,
|
| 1261 |
+
"eval_runtime": 110.8291,
|
| 1262 |
+
"eval_samples_per_second": 4.511,
|
| 1263 |
+
"eval_steps_per_second": 4.511,
|
| 1264 |
+
"step": 570
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.92,
|
| 1268 |
+
"grad_norm": 5.003520965576172,
|
| 1269 |
+
"learning_rate": 3.6426448978995054e-07,
|
| 1270 |
+
"loss": 3.6271,
|
| 1271 |
+
"step": 575
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.928,
|
| 1275 |
+
"grad_norm": 4.7706403732299805,
|
| 1276 |
+
"learning_rate": 2.9910592850826983e-07,
|
| 1277 |
+
"loss": 3.7089,
|
| 1278 |
+
"step": 580
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.928,
|
| 1282 |
+
"eval_loss": 3.6552963256835938,
|
| 1283 |
+
"eval_runtime": 111.738,
|
| 1284 |
+
"eval_samples_per_second": 4.475,
|
| 1285 |
+
"eval_steps_per_second": 4.475,
|
| 1286 |
+
"step": 580
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 0.936,
|
| 1290 |
+
"grad_norm": 5.243898391723633,
|
| 1291 |
+
"learning_rate": 2.402755304856974e-07,
|
| 1292 |
+
"loss": 3.6502,
|
| 1293 |
+
"step": 585
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 0.944,
|
| 1297 |
+
"grad_norm": 5.9423298835754395,
|
| 1298 |
+
"learning_rate": 1.8781167242860276e-07,
|
| 1299 |
+
"loss": 3.6603,
|
| 1300 |
+
"step": 590
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 0.944,
|
| 1304 |
+
"eval_loss": 3.656373977661133,
|
| 1305 |
+
"eval_runtime": 110.7007,
|
| 1306 |
+
"eval_samples_per_second": 4.517,
|
| 1307 |
+
"eval_steps_per_second": 4.517,
|
| 1308 |
+
"step": 590
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 0.952,
|
| 1312 |
+
"grad_norm": 4.393427848815918,
|
| 1313 |
+
"learning_rate": 1.4174857797209951e-07,
|
| 1314 |
+
"loss": 3.6765,
|
| 1315 |
+
"step": 595
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 0.96,
|
| 1319 |
+
"grad_norm": 4.435468673706055,
|
| 1320 |
+
"learning_rate": 1.0211629535501811e-07,
|
| 1321 |
+
"loss": 3.6715,
|
| 1322 |
+
"step": 600
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.96,
|
| 1326 |
+
"eval_loss": 3.6557369232177734,
|
| 1327 |
+
"eval_runtime": 111.0884,
|
| 1328 |
+
"eval_samples_per_second": 4.501,
|
| 1329 |
+
"eval_steps_per_second": 4.501,
|
| 1330 |
+
"step": 600
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.968,
|
| 1334 |
+
"grad_norm": 4.849489212036133,
|
| 1335 |
+
"learning_rate": 6.894067781860636e-08,
|
| 1336 |
+
"loss": 3.6554,
|
| 1337 |
+
"step": 605
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 0.976,
|
| 1341 |
+
"grad_norm": 6.290756702423096,
|
| 1342 |
+
"learning_rate": 4.2243366741735457e-08,
|
| 1343 |
+
"loss": 3.6473,
|
| 1344 |
+
"step": 610
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.976,
|
| 1348 |
+
"eval_loss": 3.6551318168640137,
|
| 1349 |
+
"eval_runtime": 111.5978,
|
| 1350 |
+
"eval_samples_per_second": 4.48,
|
| 1351 |
+
"eval_steps_per_second": 4.48,
|
| 1352 |
+
"step": 610
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.984,
|
| 1356 |
+
"grad_norm": 5.660240650177002,
|
| 1357 |
+
"learning_rate": 2.2041777523627018e-08,
|
| 1358 |
+
"loss": 3.6221,
|
| 1359 |
+
"step": 615
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.992,
|
| 1363 |
+
"grad_norm": 5.215294361114502,
|
| 1364 |
+
"learning_rate": 8.349088223306157e-09,
|
| 1365 |
+
"loss": 3.6463,
|
| 1366 |
+
"step": 620
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.992,
|
| 1370 |
+
"eval_loss": 3.655787944793701,
|
| 1371 |
+
"eval_runtime": 111.9272,
|
| 1372 |
+
"eval_samples_per_second": 4.467,
|
| 1373 |
+
"eval_steps_per_second": 4.467,
|
| 1374 |
+
"step": 620
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.0,
|
| 1378 |
+
"grad_norm": 5.055933475494385,
|
| 1379 |
+
"learning_rate": 1.1742309631845861e-09,
|
| 1380 |
+
"loss": 3.6125,
|
| 1381 |
+
"step": 625
|
| 1382 |
+
}
|
| 1383 |
+
],
|
| 1384 |
+
"logging_steps": 5,
|
| 1385 |
+
"max_steps": 625,
|
| 1386 |
+
"num_input_tokens_seen": 0,
|
| 1387 |
+
"num_train_epochs": 1,
|
| 1388 |
+
"save_steps": 50,
|
| 1389 |
+
"stateful_callbacks": {
|
| 1390 |
+
"TrainerControl": {
|
| 1391 |
+
"args": {
|
| 1392 |
+
"should_epoch_stop": false,
|
| 1393 |
+
"should_evaluate": false,
|
| 1394 |
+
"should_log": false,
|
| 1395 |
+
"should_save": true,
|
| 1396 |
+
"should_training_stop": true
|
| 1397 |
+
},
|
| 1398 |
+
"attributes": {}
|
| 1399 |
+
}
|
| 1400 |
+
},
|
| 1401 |
+
"total_flos": 6.8614575095808e+17,
|
| 1402 |
+
"train_batch_size": 1,
|
| 1403 |
+
"trial_name": null,
|
| 1404 |
+
"trial_params": null
|
| 1405 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb14f46e288db2bf5f3e7d704f0842073a12fcaa8518d89aec92a116e6e05c18
|
| 3 |
+
size 5368
|